PhD/Postdoc Fire Dynamics Modelling using Machine Learning Methods
The official announcement (german / english) is provided here: Recruitment System BUW. In the following, an unofficial summary of the english version of the Job-ID 22709 is given.
This position is to be filled as soon as possible for 3 years with 75 % (PhD) or 100% (Post Doc) of the tariff working hours (part-time is possible, please state when applying whether you would also be interested in a part-time job).
We are an interdisciplinary and international team at the University of Wuppertal (BUW) and are looking for support for our research work in the field of computer-aided modelling in civil engineering, especially fire dynamics. This position is intended to expand our methodological spectrum to include approaches from machine learning. Support us in developing more efficient and precise fire spread models!
In numerical fire simulations, design fires are often used to determine the dynamics of the fire. However, this cannot take into account local conditions such as ventilation, heat interaction or material properties during fire propagation. Only by using a complex pyrolysis model, the dynamics of fire spread can be simulated self-consistently. The challenge here is that many material parameters are required, most of which are not known or directly measurable. By means of small-scale experiments, such as a thermogravimetric analysis or a cone calorimeter, in conjunction with inverse modelling, these parameters can be calculated with the help of supercomputers.
Within the framework of the BMBF-funded BESKID project, BUW is investigating new approaches based on artificial intelligence. These will be used to determine material parameters directly from small-scale experiments. The new model will provide material parameters for simulations of fire propagation, which will be used by the project consortium for validation and application studies. The project is being carried out in collaboration with industrial partners, the Research Centre Jülich (FZJ) and the Chair Fire Dynamics at BUW. The method development is based on small-scale laboratory experiments conducted at the FZJ. Thus, a close scientific exchange between the BUW and the FZJ is aimed at. In cooperation with the industrial partners, the methods are used for testing and validation in real-scale experiments in the application case of rail transport vehicles and compared with classical approaches.
Professional and personal requirements:
- Completed university studies (Master’s degree or comparable) in engineering, physics or a related field, desirably with a doctorate in the field of fire dynamics
- Expertise in the conception and implementation of scientific simulation models, as well as in the field of machine learning
- Experience in programming in Python or C++
- Willingness to work in an interdisciplinary and international team
- Strong interest in complex issues and basic research
- Good social and communication skills
- Very good knowledge of the English language
Responsibilities and duties:
- Development and application of computer models in fire dynamics
- Evaluation of experimental data for model validation
- Cooperation in national and international collaborations
- Participation in project meetings
- Presentation of results at meetings and conferences
- Preparation of interim and final reports
- Further training in the acquisition, processing and management of research projects, integration into personnel management and university teaching
- Embedding of research activities in an interdisciplinary and international research environment
- Opportunities for flexible working hours, support with childcare and suitable solutions for your individual life situation
Applications including all relevant credentials (motivation letter, CV, proof of a successful graduation and/or PhD, job references, letters of reference, links to software projects, proof of a severe disability) should be addressed to Prof. Dr. Lukas Arnold and solely submitted via the online portal of the University of Wuppertal. Kindly note, that incomplete applications will not be considered.
Application deadline: 08.01.2023